María Villarroya Gaudó and Darío Suárez Gracia, Professors of Computer Architecture and Technology at the Engineering and Architecture School from the University of Zaragoza, and members of the Computer Architecture Research Group, gaZ, at the same university.
In this talk, we’ll introduce our research group at the University of Zaragoza, one of the universities, along with UVT, in the UNITA European Alliance. From a practical point of view, we’ll present the group’s activities and opportunities, focusing on the current work we’re carrying out.
On one hand, Professor Villarroya Gaudó will discuss opportunities to improve biomedical applications once code is optimized for specific hardware. This allows for better results and new simulation scenarios that can help solve various problems.
On the other hand, Professor Suárez Gracia will comment on two research lines of the group in two big areas: AI for greener systems and greener systems for AI. On the former, the focus will be how neural networks can improve load balancing in CPU plus GPU heterogeneous systems. On the later, the topic will move to greener systems for AI showing how to accelerate Bayesian Neural Networks on RISC-V edge-oriented cores.
Short Bio:
María Villarroya
María Villarroya Gaudó holds a degree in Physics from the University of Zaragoza and completed her PhD in Electronic Engineering at the Autonomous University of Barcelona. She has been a full-time lecturer at the University of Zaragoza since the 2003/2004 academic year, becoming a Senior Lecturer in 2020. Her research career currently focuses on heterogeneous multiprocessor systems, with a special interest in workload balancing. Previously, she focused on the analysis and design of on-chip interconnection networks for performance improvement and power saving in multiprocessor systems, and on modeling and synthesis of CMOS devices for memory element applications. She also researches new technologies applicable beyond CMOS in computer architectures. This work has been carried out within the Computer Architecture research group of the Aragon Engineering Research Institute at the University of Zaragoza.
In her teaching experience, she has been responsible for a large number of core and compulsory subjects in Computer Engineering at both Bachelor and Master levels. She has also co-supervised PhDs and has experience in management positions at the University. She is currently the director of the UNITA office at the University of Zaragoza.
Darío Suárez Gracia
Darío Suárez Gracia received his PhD degree in Computer Engineering from the University of Zaragoza, Spain, in 2011. From 2012 to 2015, he was with Qualcomm Research Silicon Valley working on programming models for mobile SoC. He was on the team that proposed the Snapdragon Heterogeneous Compute SDK. He was also the lead engineer for the Snapdragon Power Optimization SDK, which firstly demonstrated control of performance and power efficiency on mobile SoC. Additionally, he contributed to techniques used in Qualcomm’s flagship products.
Currently, he is a full professor at the University of Zaragoza. His research interests include parallel programming, heterogeneous computing, memory hierarchy design, energy-efficient multiprocessors, and fault-tolerance. He is a member of the Aragon Institute of Engineering Research (I3A), the Spanish Society of Computer Architecture (SARTECO), the IEEE Computer Society, ACM, and the HiPEAC European NoE.
